Diabetes Mellitus
Haitham S. Abu-Lebdeh
Gregg D. Simonson
TYPE 1 DIABETES MELLITUS
Definition
Type 1 diabetes mellitus (DM) is a chronic metabolic disorder of glucose homeostasis that manifests secondary to absolute lack of insulin.
Etiology
It is thought that, in humans, environmental factors such as diet, severe stress, and possibly viral infections, among other unknown factors, may trigger a T cell-mediated autoimmune destruction of pancreatic beta cells in a susceptible host, which leads to onset of type 1 DM. Family studies failed to identify a specific mendelian pattern of inheritance for this disease [1, 2]. However, multiple genetic loci are strongly linked to the development of this polygenic disorder, especially human leukocyte antigen (HLA)-DR and HLA-DQ alleles of the histocompatibility complex. Having a specific genotype that is associated with type 1 DM does not necessarily result in development of this disease; in more than half of monozygotic twins of patients with diabetes, type 1 DM will not develop, which suggests an important role for environmental factors in the etiology of this disease. Furthermore, most patients (85%) lack a family history of a similar disorder.
Genetics
Major Histocompatibility Complex Genes
Major histocompatibility complex (MHC) class II molecules attach to exogenous peptides and then present these peptides on the cell surface for T-cell (CD4) recognition. In humans, class II loci (HLA-DR, HLA-DP, HLA-DQ) of the MHC class II are located on chromosome 6. The finding that type 1 DM develops in 30% to 50% of monozygotic twins, whereas it occurs in only 15% of HLA-identical sibs, indicates that although MHC genes are strongly associated with increased risk for type 1 DM, other genes must be involved in the etiology of this disease. MHC class I alleles are also associated with type 1 DM although with a lesser effect.
Insulin-Dependent Diabetes Mellitus Genes
Multiple other genes are suspected in the development of type 1 DM. These genes are termed insulin-dependent DM (IDDM) genes. IDDM1 is the HLA region locus mentioned previously. IDDM 2 is a nonhistocompatibility gene located on chromosome 11, which contains the insulin gene. Changes in the promoter region of the insulin gene increase the risk of type 1 DM. In addition, other genes have been associated with type 1 DM; these include protein tyrosine phosphatase, nonreceptor type 22 (PTPN22) gene on chromosome 1, interleukin-2 receptor alpha gene (IL2RA), cytotoxic T-lymphocyte antigen 4 (IDDM 12, which lies on chromosome 2), interferon induced with helicase c domain 1 (IFIH1), vitamin D receptor (VDR), and others [3].
Rare Forms
Type 1 DM is rarely caused by single gene mutation; examples include the immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX), and the autoimmune polyendocrine syndrome type 1.
Epidemiology
Based on National American Health Service surveys, the prevalence of diabetes among people younger than 20 years is around 2 per 1,000 in the United States. Worldwide incidence and prevalence rates vary significantly, depending on the population, ethnicity, and geography, which suggest an important role for environmental factors. In the United States, the overall incidence of type 1 DM is around 20 per 100,000 per year (Rochester, MN, data 1970-1979). Incidence of type 1 DM is two to three times more common in whites than in other ethnic groups in the United States. Interestingly, incidence rates of type 1 DM are increasing worldwide.
Type 1 DM also develops in adults; the incidence is estimated around 8.2 per 100,000 annually [4].
Pathophysiology
The development of type 1 DM starts with an unknown precipitating event in a genetically susceptible host (stage 1) that triggers a T cell-mediated autoimmune destruction of beta cells (stage 2). Over time, generally months to years, progressive loss of insulin is noted and can be detected by using intravenous glucose tolerance tests (stage 3). Subsequently, blood glucose starts to increase, indicating significant beta cell damage (stage 4); this stage may last for several months in children or a longer period in adults and is characterized by clinical diabetes in the presence of normal or low C-peptide levels. Finally, in stage 5, insulin and C-peptide production ceases, and the subject becomes dependent on exogenous insulin for survival.
Diagnosis
Screening
Diabetes develops in 2% to 5% of relatives of patients with type 1 DM. Numerous antibodies directed against islet cell antigens have been identified, but few are of value as research screening tools [5]. In the research setting, autoantibodies are commonly used to predict type 1 DM in studies involving relatives of patients with diabetes [6]. Using low titers as diagnostic values predicts future development of type 1 diabetes in a highly sensitive manner, but this, of course, is associated with a significant number of false-positive results. Therefore, researchers use combination autoantibodies with the intravenous glucose tolerance test to increase the specificity and yield of these tests.
Serum Antibodies
The presence of serum islet cell cytoplasmic antibodies (ICAs) is a highly predictive marker for future development of type 1 DM in relatives of family members with that disease. The sensitivity is higher than 80%, and the specificity is higher than
90%. However, these values change significantly depending on the cutoff titers used. The exact antigenic molecules responsible for ICAs are not fully identified [7]. The sensitivity of Glutamic Acid Decarboxylase (GAD) antibodies and ICA 512 is between 65% and 70% and is highly specific (typically >90%). GAD is expressed in the islets of Langerhans and also in neurons and gonadal tissue.
90%. However, these values change significantly depending on the cutoff titers used. The exact antigenic molecules responsible for ICAs are not fully identified [7]. The sensitivity of Glutamic Acid Decarboxylase (GAD) antibodies and ICA 512 is between 65% and 70% and is highly specific (typically >90%). GAD is expressed in the islets of Langerhans and also in neurons and gonadal tissue.
Insulin autoantibodies (IAAs) are less sensitive markers than ICAs (25% sensitivity); however, IAAs correlate well with the short duration to development of disease as well as young-age onset of disease. Antibodies against zinc transporter antigen 8 (ZnT8) were detected in previously classified type 1 negative autoantibody status. Risk for diabetes increases with higher titers of ICA and the presence of multiple autoantibodies. Autoantibodies may be of value to screen the high-risk population, but lack of an effective intervention is the main barrier against recommendation of such a practice.
First-Phase Insulin Release
Measuring first-phase insulin release during an intravenous glucose tolerance test increases the predicted value of screening with autoantibodies [8]. First-phase insulin release is consistent with progress of diabetes from stage 2 to stage 3. A finding of insulin release that is below the first percentile confers a 90% risk of type 1 diabetes in 3 years in patients with positive ICAs and positive IAAs.
Monogenic Forms of Diabetes Syndromes
Sometimes the diagnosis of type 1 diabetes from other rare diabetes syndromes is not clear from autoantibody studies. Recent advances in commercially available genetic testing allow the distinction of type 1 diabetes from the rather rare monogenic forms of diabetes that include maturity-onset diabetes of the young (MODY) and neonatal diabetes. Genetic screening for monogenic diabetes should be considered in the following clinical scenarios: diabetes diagnosed within 6 months of birth and in children diagnosed with diabetes without autoantibodies (indicative of type 1) or signs of insulin resistance (obesity, acanthosis nigricans indicative of type 2 diabetes).
Clinical Manifestations
The most common presenting symptoms are polydipsia and polyuria. Other symptoms frequently encountered in children and adolescents are fatigue, weight loss, and abdominal pain, as well as enuresis. Diabetic ketoacidosis (DKA) is a presenting symptom in 10% to 40% of cases. Frequency of this presentation varies significantly between different countries and may depend on public education, incidence of diabetes in that region, and presence of other people with diabetes in the family, resulting in early detection before development of DKA. These symptoms are not specific for type 1 diabetes and can occur in patients with type 2 diabetes, including DKA.
Most patients with type 1 diabetes are diagnosed at a young age (typically, less than 30 years); furthermore, in recent years, an increasing number of adolescents have been diagnosed with type 2 DM, especially those who are obese.
Laboratory Findings
The laboratory glucose values used for diagnosis type 1 DM are fasting plasma glucose of 126 mg/dl (7 mmol/l) on more than two occasions or 2-hour glucose value of 200 mg/dl (11.1 mmol/l) or higher during oral glucose tolerance testing. Recently in 2010, glycosylated hemoglobin A1c (A1C) of greater than or equal to 6.5% has been accepted by American Diabetes Association (Clinical Practice Recommendations 2010) as diagnostic for diabetes.
Prevention
Prevention of type 1 DM is still in early research stages [9, 10]. Currently, no effective preventive strategies exist. Immunomodulating drugs (e.g., azathioprine
and cyclosporine) and nicotinamide prevented further beta cell destruction in small studies for a short time. The short-term benefits were not sufficient to indicate long-term use of these potentially toxic medications. The Diabetes Prevention Trial (DPT-1) demonstrated that using injectable or oral insulin does not prevent diabetes in relatives of patients with diabetes. Ongoing studies are evaluating the effect of oral insulin on subpopulations of relatives of patients with diabetes.
and cyclosporine) and nicotinamide prevented further beta cell destruction in small studies for a short time. The short-term benefits were not sufficient to indicate long-term use of these potentially toxic medications. The Diabetes Prevention Trial (DPT-1) demonstrated that using injectable or oral insulin does not prevent diabetes in relatives of patients with diabetes. Ongoing studies are evaluating the effect of oral insulin on subpopulations of relatives of patients with diabetes.
Treatment
Insulin
Only insulin is used to control blood glucose in type 1 diabetes. Animal-derived insulins have long been passed over in favor of genetically engineered so-called human insulins. The goal of treatment is to achieve near-normal plasma glucose around the clock. Intensive therapy delays the onset and progression of microvascular complications of diabetes but may not affect macrovascular mortality [11, 12]. The main disadvantages of intensive therapy are the increased frequency of hypoglycemia, ketoacidosis, and weight gain [13]. Ideally, reducing the hemoglobin A1c to less than 7% (ADA) is the goal in type 1 DM patients [14]. Patients are typically educated about recognizing and treating hypoglycemia at the same time they learn about using insulin.
Insulin Regimens
Conventional Therapy. The regimen of twice-daily injections with basal insulin (typically intermediate-acting insulin) and a bolus insulin (regular or rapidacting analogues)—known as conventional therapy—is associated with variable and inferior glycemic control in comparison with intensive insulin therapy regimens and therefore has been abandoned in favor of other insulin regimens. Evidence supports intensive glycemic control by using multiple daily injections or insulin pump rather than two-injection conventional therapy [15]. A twicedaily split insulin regimen (conventional therapy) and once-daily injections are not recommended because of difficulty in achieving the goal and the inability to adjust insulin doses appropriately to prevent hyperglycemia in type 1 DM. These two modalities of treatment, however, are frequently used by general practitioners at the onset of disease until the patient is transferred to the care of a specialist.
Intensive Insulin Therapy. Insulin pump: Portable, external insulin infusion pumps have undergone significant improvement in the past two decades [16]. Insulin is delivered continuously at a set basal rate (˜60% of total daily insulin). Pumps can be programmed to increase the basal insulin at night to counteract morning hyperglycemia. At mealtime, bolus doses of insulin are provided to cover meal glucose excursions. Use of such a pump requires frequent monitoring of blood glucose and predisposes to site infections and DKA but usually results in better glycemic control.
Multiple Daily Injections: Multiple daily injections (MDIs) refer to three or more injections with very-rapid-acting analogues before meals. In addition, basal insulin is provided at dinner (glargine or detemir) or at bedtime (NPH or glargine or detemir) for background coverage over a 24-hour period. Frequently, if NPH or detemir is used, a second dose is provided before breakfast. In type 1 DM, intensive insulin therapy (pumps or MDIs) provides lower hemoglobin A1c and significantly reduces the risk of microvascular complications. These data are derived primarily from the results of the Diabetes Control and Complications Trial (DCCT/EDIC) cohort, which also involved a comprehensive patient support program of diet, exercise, and close supervision with instructions.
Table 6.1. Insulin Types and Timeframe of Action | ||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
Types of Insulin
Very-rapid-acting insulin analogues. The available agents, insulin aspart (NovoLog), insulin glulisine (Apidra), or insulin lispro (Humalog), are engineered so that after injection, the insulin dissociates quickly from the aggregate. Therefore, these insulins have rapid onset of action and short duration of activity (Table 6.1). These agents are used specifically to lower glucose after a meal and to correct postprandial hyperglycemia and therefore are called “meal” insulins. These insulins are at least as effective as regular insulin [15, 17, 18, 19]. Furthermore, their use can reduce the frequency of hypoglycemia and can be safely used in patients with unpredictable eating patterns.
Regular insulin. The main disadvantage for use of regular insulin is the need to inject it 30 to 45 minutes before meals, which may be inconvenient or may be associated with hypoglycemia if the meal is delayed or not eaten.
Intermediate-acting insulin. Neutral protamine Hagedorn (NPH) insulin has a longer duration of action than regular insulin and is used mainly to provide basal insulin coverage [20]. NPH insulin does not work fast enough to control postprandial glucose level right after administration but may be used to over a meal 4 to 5 hours after injection. When administered at bedtime, NPH is as effective in reducing fasting plasma glucose as long-acting insulin analogs, but increases the risk of nocturnal hypoglycemia [3].
Long-acting insulin analogues. Glargine (Lantus) and detemir (Levemir) are long-acting synthetic preparations that are relatively peakless, with a lower incidence of hypoglycemia and a prolonged duration of action. The main disadvantage is that they cannot be mixed with other insulins. Detemir insulin, although has a shorter half-life than glargine, produces similar effects to glargine when administered properly.
Premixed insulins. Mixtures of two kinds of insulin do not allow for flexibility and require more skill in adjustment of insulin to achieve a therapeutic end, and thus, such a combination should not be the drug of choice.
Jet injectors have been used with success by patients or caregivers that are unable to use a standard syringe/needle technique.
Pramlintide
Pramlintide: Pramlintide (15-60 µg) is a synthetic analogue of human amylin that slows gastric emptying and lowers A1c concentrations mainly by reducing postprandial glucose excursions. In patients with type 1 diabetes, it is used only with meal time insulins. It is recommended to reduce the insulin dose by 50% when pramlintide is started.
Inhaled insulin: The first commercially available inhaled insulin (Exubera) was pulled from the market due to poor acceptance by patients and health professionals. One formulation of very-rapid-acting inhaled insulin is still undergoing clinical development and whether this is approved by the Food and Drug Administration and becomes commercially available has yet to be determined.
Diet
Dietary management regimens improve glycemic control, but insufficient evidence is available to recommend a specific diet plan over another. Dietary knowledge is essential for carbohydrate counting if used in patients using MDIs or insulin pumps. Eating disorders are more common in type 1 DM, especially in adolescents, and this adversely influences glycemic control. Therefore, regular psychological assessment and instruction regarding healthy eating habits are recommended [22].
Exercise
Aerobic exercise is recommended. In addition, persistent training with light weights and high repetitions could be useful [23]. Of note, exercise during a state of insulin deficiency as manifested by higher blood glucose before the activity may be associated with hyperglycemia subsequent to exercise, and therefore, blood glucose monitoring is helpful, especially in cases of unanticipated exercise [24]. Insulin should not be injected into the exercising limb because continuous muscle movement results in increased insulin release. The abdomen is the preferred site for insulin injection.
High-intensity exercise may result in increased albuminuria and may in theory be associated with adverse effects. However, no evidence supports clinical progression of retinopathy or kidney disease with high-intensity exercise. The risk of myocardial infarction (MI) is higher in people with type 1 DM, and thus, the American Diabetes Association recommends cardiac stress exercise testing for patients who are older than 35 years, who have been diagnosed with type 1 diabetes for longer than 15 years (especially those who have evidence of clinical autonomic neuropathy, peripheral vascular disease, or microvascular disease), or who have a significant cardiovascular risk profile. The evidence to support such recommendations is inadequate.
Blood Glucose Monitoring
Learning blood glucose self-monitoring skills is essential for type 1 DM. It is important for patients with diabetes to monitor for hypoglycemia and significant hyperglycemia to avoid complications. Little evidence indicates that frequent blood glucose self-monitoring translates to better glucose control by using standard glucometers. This might be secondary to lack of adjustment or intervention by physicians or patients. The use of laser blood glucose monitoring devices, home glycated hemoglobin monitors, fructosamine monitors, and measurement of advanced glycation end products by skin autofluorescence has not been shown to provide advantage over standard methods. Blood glucose monitors with electronic voice and integrated lancing-blood sample monitors are effective and helpful in blind patients. Continuous-monitoring real-time glucose sensors are helpful in adjusting insulin doses and are used typically over a period of 72 hours to obtain data. Glucose measurements using these devices correlate very well with standard glucometer readings. Such devices in patients with type 1 or type 2 DM were superior to standard meters in reducing duration of hyperglycemia and hypoglycemia and are used for prolonged time (more than 72 hours) in patients with type 1 DM and hypoglycemia unawareness that require glucagon or serious resuscitation.
Transplantation
Pancreas transplantation is considered in patients planning to have a kidney transplant for treatment of end-stage renal disease, given that pancreas transplantation may improve the survival of the transplanted kidney, may improve hypoglycemia, and may partially reverse neuropathy. Pancreas transplantation alone may be considered in patients with severe complications of diabetes. Islet cell transplantation has a significant advantage over whole-pancreas transplantation but requires special experience that is not available in many centers. In addition, a sufficient supply of islet cells is lacking. Stem cell transplantation has been attempted in small numbers of patients with type 1 DM.
Diet
Dietary management regimens improve glycemic control, but insufficient evidence is available to recommend a specific diet plan over another. Dietary knowledge is essential for carbohydrate counting if used in patients using MDIs or insulin pumps. Eating disorders are more common in type 1 DM, especially in adolescents, and this adversely influences glycemic control. Therefore, regular psychological assessment and instruction regarding healthy eating habits are recommended [22].
TYPE 2 DIABETES MELLITUS
Definition
Type 2 DM is a chronic disorder of glucose homeostasis characterized by hyperglycemia and impaired insulin action, with abnormal pancreatic insulin secretion as well as increased rates of hepatic glucose production. Unlike type 1 DM, no absolute physiologic lack of insulin is present; rather, a state of relative insulin deficiency develops.
Etiology
The concordance rate of type 2 DM in identical twins is 70% to 90%, with a strong familial clustering of type 2 DM, suggesting a genetic etiology. No specific gene has been identified as the cause of type 2 DM, and multiple genetic abnormalities may be involved. Insulin resistance alone does not explain diabetes; impaired beta cell function manifested as impaired first- and second-phase insulin secretion combined with a decline in incretin action is also implicated. It is clear that type 2 diabetes has both environmental (associated with obesity, nutrition, and/or reduced activity) and genetic components.
Risk Factors
Overweight and Obesity. Risk for type 2 DM increases with obesity as measured by the body mass index (BMI) in both men and women [25]. Overweight is considered BMI greater than or equal to 25 kg/m2. Central fat (so-called apple distribution) increases the risk of type 2 DM in addition to BMI measurements. Central obesity is defined as a waist circumference greater than 40 inches in men and greater than 35 inches in women. Weight gain in adulthood of more than 10 kg in men or more than 8 kg in women is associated with increased risks of DM regardless of the BMI.
Ethnicity. The reasons behind ethnic variation are unclear, but general themes were observed among minorities at increased risk for diabetes (e.g. Pima indians and micronesian Nauru). These include abandoning traditional lifestyle behaviors and adopting new behaviors that include reduced physical activity and increased caloric intake.
Family history of type 2 DM, especially in first degree relatives
Patients with elevated fasting glucose measurements (100-125 mg/dl) or with high postprandial measurements (2 hr OGTT value 140-199 mg/dl) and/or A1C 5.7% to 6.4%.
Lack of exercise or physical inactivity. This is an independent risk factor from the BMI.
Dyslipidemia (HDL < 35 mg/dl and/or triglycerides >250 mg/dl)
Hypertension (>140/90 mmHg) or treated for high blood pressure
History of gestational diabetes mellitus (GDM) or baby weight greater than 9 lbs (4 kg)
Syndromes associated with insulin resistance (severe obesity, polycystic ovary syndrome and/or acanthosis nigricans)
Epidemiology
Based on US national health surveys from the Centers for Disease Control and Prevention, it is estimated that approximately 26.4 million in the United States have diabetes that equals approximately 8.3% of the population. The prevalence of diabetes increases with age, and 20% of subjects older than 65 have diabetes. Worldwide, the prevalence of diabetes differs significantly between one region and another, with some areas having extremely high prevalence (e.g., Micronesian Anurans rates ˜40%).
Diabetes is the sixth leading cause of death in the United States, and most deaths are attributed to heart disease. The American Diabetes Association estimates health care costs that are specifically due to diabetes (direct medical costs) at $116 billion in 2007, plus another $58 billion in indirect costs of disability, work loss, and premature mortality.
Pathophysiology of Type 2 Diabetes Mellitus
Three primary pathophysiologic features of type 2 DM are insulin resistance, relative insulin deficiency, and impaired incretin action. Insulin resistance in muscle, liver, and adipose tissue is related to weight gain. Through complex perturbations in insulin signaling pathways and fuel metabolism (beyond the scope of this chapter), these tissues become resistant to insulin. Early on in the progress from normal glycemia to prediabetes and eventually type 2 DM, the pancreas responds by secreting excessive insulin to overcome the insulin resistance. With time, the pancreas cannot keep up with the demand, and a state of relative insulin deficiency occurs marked by elevation in both fasting BG, due to improper suppression of hepatic gluconeogenesis, and postmeal hyperglycemia. Many factors lead to the decline in insulin secretion including a loss of first-phase insulin secretion and a blunted second phase. Gut hormones like glucagon-like peptide-1 and glucose-dependent insulinotropic peptide improve insulin secretion and beta cell function/health, and as diabetes develops, there level or action in the body declines partially, causing the decline in beta cell function. Other factors may include severe glucotoxicity and lipotoxicity that may impede normal beta cell function. An important feature of type 2 DM is that the disease is progressive in nature, and insulin resistance tends to remain high throughout the disease, and beta cell function continues to decline.
Diagnosis
Clinical Manifestations
Unlike patients with type 1 DM, most patients with type 2 diabetes do not show the classic symptoms of hyperglycemia, as mentioned previously. Recent estimates by the Centers for Disease Control and Prevention (CDC) estimate 7 million patients in the United States are undiagnosed with DM. The most common classic symptoms in type 2 DM are excessive thirst followed by easy fatigability, neurologic symptoms, blurred vision, and/or recurrent infections.
Laboratory Findings
Plasma glucose levels that are thought to be, in the long term, associated with retinopathy and proteinuria are used to diagnose diabetes. Several diagnostic modalities have been proposed to separate those patients with hyperglycemia who are at increased risk for development of microvascular complications from those who are at low risk but who do have hyperglycemia. Currently, two main diagnostic criteria are used: the American Diabetes Association criteria, which use fasting plasma glucose, and the World Health Organization criteria, which depend on oral glucose tolerance test results. Various populations were followed up prospectively by using a 2-hour glucose test after a 75-g glucose load for the development of microvascular complications. A postchallenge plasma glucose level of 200 mg/dl (11.1 mmol/l) seemed to differentiate reliably subjects who experience microvascular complications from those who do not. These studies are the basis of the World Health Organization diagnostic system.
In 2010, the American Diabetes Association added glycosylated hemoglobin (A1c) greater than or equal to 6.5% to its list of diagnostic criteria for type 1 and type 2 DM. With the standardization of the A1C assay, it has become a widely accepted means to diagnosis diabetes. The ADA retains the other three diagnostic criteria of a fasting plasma glucose level of greater than or equal to 126 mg/dl (7 mmol/l), random plasma glucose greater than or equal to 200 mg/dl (11.1 mmol/l) plus symptoms, and 75-g oral glucose tolerance test 2-hour value of greater than or equal to 200 mg/dl (11.1 mmol/l) as acceptable diagnostic criteria for DM. All diagnostic tests should be confirmed on a subsequent day unless signs of unequivocal hyperglycemia.
Screening
Identifying patients with asymptomatic diabetes is an effective strategy because of the availability of effective treatments that reduce the morbidity and the progression of disease. The American Diabetes Association recommends screening overweight adults 45 years or older every 3 years as the consensus. Opportunistic screening patients younger than 45 years old with multiple risk factors is also encouraged because the incidence and prevalence of type 2 DM are increasing in children, adolescents, and young adults. Risk factors (described in detail above) include family history of diabetes, overweight defined as BMI of 25 kg/m2 or higher, habitual physical inactivity, being a member of a high-risk ethnic or racial group, previously identified impaired fasting glucose or impaired glucose tolerance, hypertension, dyslipidemia, history of gestational DM or delivery of a baby weighing more than 9 lb (4 kg), polycystic ovary syndrome, and/or acanthosis nigricans.
Prevention
The most effective recommendation to prevent diabetes is to engage in lifestyle modifications that lower caloric intake and increase physical activity, ultimately resulting in modest weight loss (5%-7% body weight). Metformin is increasingly being used off-label to prevent diabetes. Data from the Diabetes Prevention Study (DPP) revealed that metformin was most effective in people with BMI greater than 35 kg/m2 and was ineffective at preventing diabetes in lean patients. Note that lifestyle intervention is effective at all BMIs to prevent diabetes and has shown to be an effective prevention strategy for all people at risk of diabetes. Angiotensin converting enzyme inhibitors (ACE inhibitors) and angiotensin receptor blockers (ARBs) have been associated with diabetes prevention as secondary end points; however, in prospective trial, they have not shown to be effective at preventing diabetes [10].
Treatment
Weight Reduction and Dietary Restrictions
The effect of diet on glucose control is observed early, before any demonstrable weight loss. Weight loss itself is also associated with significant improvement in glycemic control, whether it is through diet and exercise, behavioral modification, weight loss medications, or bariatric surgery [27]. Programs that are associated with 5% to 10% weight loss in 3 to 4 months result in significantly improved glycemic control. Larger weight loss percentage may result in normalization of fasting glucose in newly diagnosed patients with diabetes. In the UK Prospective Diabetes Study (UKPDS), patients with mild fasting hyperglycemia were more likely to normalize glucose measurements than were patients with higher fasting glucose measurements [28]. Using a very-low-calorie diet (800 kcal/d) produces greater initial improvement in glycemic control than lowcalorie diets (1,000-1,200 kcal/d), but no measurable difference exists between these diets in the long term at 6 to 12 months. Using Orlistat (120 mg t.i.d.) in patients with type 2 DM is associated with modest weight loss but also with a reduced need for insulin or oral diabetes medicines and improved A1c (0.5% over placebo).
Exercise
Exercise in combination with diet results in maintenance of weight loss and therefore is recommended for patients with type 2 DM [29]. Research has shown that a modest weight reduction of 5% to 7% of body weight and an increase in physical activity to 150 minutes per week improves glucose control and prevents or delays the onset of diabetes [11].
Oral Drug Therapy (Table 6.2)
Biguanides
Metformin (500-2,550 mg/d) is available in immediate-release and also extendedrelease forms. In the UKPDS, overweight patients assigned to metformin had fewer diabetes-related complications and lower mortality rates than did those using insulin [30]. Compared with the conventionally treated group (using diet), patients using metformin had a 32% risk reduction for diabetes complications, a 42% risk reduction for diabetes-related death, and a 36% reduction for all-cause mortality. Therefore, metformin is recommended as the first-line oral agent to be used in patients with type 2 DM [31]. Metformin reduces plasma glucose mainly by altering hepatic gluconeogenesis and thereby reducing hepatic glucose release. Furthermore, insulin-stimulated glucose uptake in the muscle is also enhanced by the use of metformin. Metformin does not increase insulin release, and therefore, it is not associated with hypoglycemia and does not cause weight gain. Typically, metformin reduces hemoglobin A1c by 1 to 2 percentage points. The most common side effects are gastrointestinal (diarrhea and indigestion), but lactic acidosis, a very rare side effect, is the most serious adverse event and could be fatal, especially if metformin is used in patients with renal impairment, cardiac or pulmonary failure, or sepsis. The estimated glomerular filtration rate (eGFR) cutoff to safely use metformin has not been clearly delineated; however, recent evidence has shown that metformin can be used safely in stage 3 chronic kidney disease (CKD) yet remains absolutely contraindicated in CKD stages 4 and 5 (eGFR <30 ml/min).
Sulfonylureas
Second-generation sulfonylureas are used mainly in the United States and have replaced the first-generation agents tolbutamide and chlorpropamide. All agents,
however, reduce glucose levels effectively and are comparable in efficacy (lowering hemoglobin A1c by 1%-2%). All sulfonylureas function by stimulating insulin release. Patients who fail to respond to sulfonylureas are typically thin and have low insulin levels [32]. The most common side effects are weight gain of 2 to 3 kg and hypoglycemia (1%-2%), especially in the elderly. The results of the UKPDS show that the use of sulfonylureas does not increase cardiovascular events or cardiovascular motility in comparison with findings in patients who are treated with diet alone [33]. Sulfonylureas are used in combination therapy with other oral agents, as well as insulin, with variable successful results [34, 35, 36, 37, 38].
however, reduce glucose levels effectively and are comparable in efficacy (lowering hemoglobin A1c by 1%-2%). All sulfonylureas function by stimulating insulin release. Patients who fail to respond to sulfonylureas are typically thin and have low insulin levels [32]. The most common side effects are weight gain of 2 to 3 kg and hypoglycemia (1%-2%), especially in the elderly. The results of the UKPDS show that the use of sulfonylureas does not increase cardiovascular events or cardiovascular motility in comparison with findings in patients who are treated with diet alone [33]. Sulfonylureas are used in combination therapy with other oral agents, as well as insulin, with variable successful results [34, 35, 36, 37, 38].
Table 6.2. Common Oral Diabetes Therapies | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Meglitinides
Repaglinide (0.5-16 mg/d) and nateglinide (60-360 mg/d) stimulate insulin release through a mechanism slightly different from that of sulfonylureas [41]. These medications have a short duration of action and therefore should be used before meals. While normally taken three times per day, meglitinides may be added for an extra meal (up to four doses per day) and should be not taken for skipped meals, reducing the risk of hypoglycemia. Meglitinides effectively reduce hemoglobin A1c by up to 1.5%. Side effects include hypoglycemia and weight gain and should not be used in combination with sulfonylureas.
Thiazolidinediones
Rosiglitazone (2-8 mg/d) and pioglitazone (15-45 mg/d) mediate their effect by binding to nuclear receptor peroxisome proliferator-activated receptor-γ and enhance tissue (muscle) sensitivity to insulin. The most common side effects are weight gain and fluid retention (edema), so these medications should be avoided in patients with congestive heart failure, especially NYHA class III and IV. The original thiazolidinedione medication, troglitazone, was withdrawn from the market because of fatal hepatic disease. In clinical trials, rosiglitazone and pioglitazone did not show evidence of hepatotoxicity. These agents are effective as monotherapeutic agents or in combination therapy and reduce A1C by 1.0 to 1.5 percentage points. They can be used in patients with renal impairment and when used with insulin leads to increased risk of fluid retention.
The thiazolidinedione class has come under intense scrutiny in terms of safety. From the cardiovascular standpoint, pioglitazone appears to be the safer alternative as shown by no increased risk of CV events that was observed in the “PROactive” study. A meta-analysis published on major studies with rosiglitazone demonstrated a significant increased risk of CV events and trend toward increased mortality resulting in a black box warning [42]. More recently, rosiglitazone was placed in a risk evaluation and mitigation strategy (REMS) program by the FDA that requires written verification by physician that pioglitazone and other glucose-lowering therapies are not effective in lowering glucose in the patient and the potential for increased risk of CV events has been discussed with the patient. The use of pioglitazone, however, is being evaluated for possible association with increased bladder cancer. FDA 2010 safety announcement indicates an ongoing review for a possible increased risk.
Dipeptidyl-Peptidase 4 Inhibitors
The dipeptidyl-peptidase 4 (DPP-4) inhibitors sitagliptin (25-100 mg/d), saxagliptin (2.5-5 mg/d), and linagliptin (5 mg/d) work by inhibiting the enzyme responsible for the breakdown of the incretins glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic peptide (GIP). The inhibition DPP-4 allows a two- to threefold increase in these incretins, resulting in increased insulin secretion and suppression of postmeal hyperglucagonemia. The DPP-4 inhibitors are very well tolerated and do not result in weight gain or increased risk of hypoglycemia
over placebo. When used as monotherapy or when added to other oral agents, DPP-4 inhibitors lower A1C up to 1 percentage point. A purported risk of pancreatitis with sitagliptin has not been substantiated in analysis of large commercial insurance databases [43]. Recently, the labeling for sitagliptin has been modified to require checking serum creatinine with corresponding creatinine clearance before initiating therapy and periodically thereafter to make sure patients with impaired kidney function are treated with the appropriate dose. No dose adjustments are required for linagliptin for patients with impaired renal and/or liver function
over placebo. When used as monotherapy or when added to other oral agents, DPP-4 inhibitors lower A1C up to 1 percentage point. A purported risk of pancreatitis with sitagliptin has not been substantiated in analysis of large commercial insurance databases [43]. Recently, the labeling for sitagliptin has been modified to require checking serum creatinine with corresponding creatinine clearance before initiating therapy and periodically thereafter to make sure patients with impaired kidney function are treated with the appropriate dose. No dose adjustments are required for linagliptin for patients with impaired renal and/or liver function
α-Glucosidase Inhibitors
Acarbose (75-300 mg/d) and miglitol (75-300 mg/d) inhibit α-glucosidase activity in the luminal intestinal brush border, leading to delay in absorption of carbohydrates and therefore producing a reduction in postprandial glucose concentrations [39, 40]. These agents are frequently associated with bloating, flatulence, and diarrhea and therefore are not commonly prescribed. Typically, A1C decreases of ˜0.5 percentage points without significant weight gain or hypoglycemia.
Bromocriptine
Bromocriptine, a medication used to treat hyperprolactinemia and galactorrhea for many years, has been approved for the treatment of type 2 diabetes [19]. Bromocriptine is a dopamine D (2) receptor agonist that has been shown to lower insulin resistance through a not completely understood mechanism (possibly through improving circadian rhythm). A rapid-release formulation (bromocriptine QR) has been developed that is taken in the morning and results in a modest reduction of A1C of up to 0.5 percentage points with no increased risk of hypoglycemia over placebo. Common side effects of bromocriptine QR are transient nausea and diarrhea. While not commonly used, where this class of medications fits in the scheme of type 2 diabetes is still evolving.
Colesevelam
Colesevelam (3.8-4.4 g/d), a bile-acid sequestrant used to treat primary hypercholesterolemia, is also approved to treat type 2 DM. While the glucose-lowering mechanism of action is not clearly delineated, it may work by inhibiting glucose production by the liver (gluconeogenesis) and may also stimulate incretin release. Addition of colesevelam to commonly used oral agents such as metformin or sulfonylurea results in an approximately 0.5 percentage point reduction in A1C in addition to modest low-density lipoprotein (LDL) lowering. The most common side effects are constipation and dyspepsia.
Noninsulin Injectable Therapies (Table 6.3)
Exenatide
Exenatide is GLP-1 mimetic (5-10 µg b.i.d.), discovered from the salivary secretions of the venomous Gila monster that is a GLP-1 receptor agonist. The initial dose of 5 µg exenatide is injected subcutaneously within 1 hour of the morning and evening meals and is titrated to 10 µg b.i.d. after 1 month based on the level of glycemic control and tolerability. Exenatide causes an increase in glucose-dependent insulin secretion and suppression of postmeal glucagon secretion that flattens postmeal glucose excursions, resulting in approximately 1 percentage point reduction in A1C [44]. Transient gastrointestinal side effects are very common. Most patients lose weight (3-5 kg typical response) with exenatide due to enhanced satiety, and there is no increased risk of hypoglycemia. Exenatide results in modest improvement in lipids and blood pressure that accompanies weight loss. Exenatide is usually used in combination with one or two oral diabetes therapies and was recently approved by the FDA for use with long-acting insulin (glargine).
Table 6.3. Injectable Noninsulin Diabetes Therapies | ||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||
Liraglutide
Liraglutide is a GLP-1 receptor agonist (1.2-1.8 mg/d) that was designed to have a longer half-life of approximately 13 hours after subcutaneous injection. The addition of a 16-carbon palmitic fatty acid to GLP-1 creates an analog that binds with albumin in the bloodstream, protecting it from degradation by DPP-4 enzyme. Because of the longer half-life compared to exenatide, it can be injected one time per day at a consistent time before or after meals. When compared head-to-head with exenatide, it provides similar modest weight reduction (˜3 kg) and slightly better A1C lowering [45]. This might be due to the improved fasting BG reduction found with the longer-acting GLP-1 receptor agonist. Like other GLP-1 agonists, transient gastrointestinal side effects are common. Liraglutide does not increase risk of hypoglycemia over placebo when used in monotherapy or other non-hypoglycemia-induced therapies such as metformin or thiazolidinediones. The medication has been found to cause C-cell tumors or MTC in rats so should not be used in patients with medullary thyroid carcinoma or MEN 2. Off-label use of liraglutide as a weight loss medication has been studied yielding modest weight loss in patients with prediabetes and those that are obese with normal glycemia [25].
Insulin
See the section on type 1 DM regarding types of insulins. Insulin therapy should not be delayed in patients with type 2 DM, especially if glycemic control is suboptimal with the use of noninsulin agents. The primary adverse effects are weight gain (4 kg) and hypoglycemia. No significant evidence indicates that exogenous insulin use is associated with cardiovascular events. In the UKPDS, the use of insulin or sulfonylureas was not associated with increased cardiovascular disease, in comparison with findings in patients treated with diet alone.
Once-Daily Injection Regimen: Either intermediate- or long-acting insulin is used in this regimen. This regimen is associated with fewer hypoglycemic episodes but usually results in variable suboptimal glucose levels.
Twice-Daily Injection: Prebreakfast and predinner twice-daily NPH (so-called split regimen) is used in type 2 DM patients especially outside the United States. If prelunch or bedtime glucose is elevated, then very-rapid-acting insulin is added (i.e., split mixed regimen), with effective results. In some patients, this regimen can be substituted with premixed insulin, but this has the disadvantage of not allowing proper adjustments of insulin doses.
Intensive Insulin Therapy: Similar to type 1 DM, the MDI regimen is sometimes used in type 2 DM, especially if patients seek a flexible regimen, but is associated with weight gain.
Pramlintide
See section on type 1 DM. Pramlintide (60-120 µg) has been used in type 2 DM patients in combination with insulin, sulfonylureas, and metformin.
Combination Therapy
Several combination regimens of sulfonylureas, metformin, or thiazolidinediones with insulin have been investigated and are successful in reducing A1C as well as fasting plasma glucose measurements, but these combinations are more expensive than insulin alone. Data from the UKPDS demonstrate that combination therapy is need in the majority of patients to achieve glycemic target of A1C less than 7%. Combination therapy also is often needed to maintain glycemic control with long-standing type 2 diabetes due to the progressive nature of the disease. Assuming that metformin should be considered first-line therapy for type 2 diabetes, considerations for selecting medication to add to metformin include:
Does self-monitoring BG data inform whether the fasting and/or postmeal BG remains elevated? If fasting blood glucose remains elevated, then the addition
of background insulin or pioglitazone should be considered. If postmeal BG remains elevated, then incretin-based therapies or insulin secretagogues should be considered.
Weight: Does the patient seek weight loss or at least weight maintenance? Incretin-based therapies are weight neutral or result in modest weight loss. Insulin and secretagogues cause weight gain.
What is the level of patient and provider fear of hypoglycemia? Incretin-based therapies and pioglitazone do not cause hypoglycemia above placebo, and insulin and insulin secretagogues increase risk of hypoglycemia.
What is the financial status of the patient and can they afford a branded medication? Metformin, Sulfonylureas and NPH insulin provide lower cost options than branded incretin-based therapies and pioglitazone.
Transplantation
Pancreatic transplantation is not recommended for patients with type 2 DM.
DIABETIC KETOACIDOSIS
Definition
DKA is an acute, life-threatening complication of diabetes characterized by hyperglycemia, ketonemia, and a wide anion-gap metabolic acidosis. DKA occurs mainly in patients with type 1 DM and may occur in other types of diabetes [46].
Etiology
Its etiology is absolute or relative insulin deficiency in the presence of excessive counterregulatory hormones, leading to numerous metabolic abnormalities, diuresis, and ketoacid accumulation. The process could be triggered by various precipitating factors (e.g., pneumonia, urinary tract infections, and other infections; 25%) [47] or stress associated with severe or acute illness, including coronary disease, gastrointestinal hemorrhage, and trauma (10%-20%). It can also be the first presentation of diabetes in a previously undiagnosed patient (10%-30%). In insulin-requiring patients with diabetes, omission of insulin or nonadherence to therapy or suboptimal dosing preoperatively or postoperatively may precipitate DKA (30%). In patients using insulin pumps, DKA occurs because of catheters that are dislodged or obstructed [13]. Finally, medications, especially those that increase insulin resistance (e.g., glucocorticoids, β-agonists, and sympathomimetics), can precipitate DKA.
Epidemiology
The incidence of DKA seems to be increasing. In the United States, the incidence of hospital admission increased from 4 per 1,000 to 12 per 1,000 from 1980 to 1989. Generally speaking, DKA occurs in patients with type 1 DM, but patients at high risk for DKA include those at extremes of age, those with poor prior glycemic control, and those who use insulin pumps [48, 49]. The overall morality rates are about 2%, it is higher for non hospitalized patients.
Pathophysiology
Elevated levels of counterregulatory hormones are necessary for the development of DKA in patients with diabetes. Glycogenolysis is enhanced especially by glucagon, catecholamines, and low insulin levels. Furthermore, glucagon excess and insulin deficiency result in enzymatic changes in the liver that ultimately shift pyruvate away from glycolysis toward the glucose synthesis. Excessive counterregulatory hormones in the absence of insulin lead to lipolysis and excess free fatty acid release from adipose tissue, which is then converted to ketoacids in the liver. Ketoacids are buffered by bicarbonate, leading to its depletion.
Diagnosis
Laboratory Findings
No specific laboratory criteria exist for the diagnosis of DKA; however, the presence of wide anion gap [(Na + K) – (Cl + HCO3) >16 mEq/l] and ketonemia in a patient with diabetes is consistent with DKA. Most patients’ glucose levels are high, exceeding 250 mg/dl (14 mmol/l), but occasionally glucose levels are normal, especially if the episode is preceded by long periods of fasting. Hematocrit and mean corpuscular volume increase in DKA. Glucose enters the red blood cell easily despite lack of insulin, which leads to osmotic swelling of the blood cell. Serum sodium levels vary significantly despite total body sodium deficit. Triglyceride levels are frequently elevated, mainly as a result of insulin deficiency and reduced clearance, but they typically return to normal with proper treatment of DKA. Initially, serum potassium levels are high or high normal secondary to the dehydration of acidosis, although total body potassium stores are depleted. Serum phosphate levels are also elevated initially, and both potassium and phosphate decrease quickly and significantly with treatment of DKA. Both amylase and lipase levels may be elevated in DKA without abdominal symptoms of pancreatitis or even hypertriglyceridemia; however, careful evaluation for possible pancreatitis should be pursued if lipase is elevated.
Clinical Manifestations
Symptoms develop rapidly in patients using pumps but may take several days in other patients with diabetes. Ketonemia results in nausea and vomiting and the characteristic odor of acetone on the breath. Less frequently, abdominal pain is the presenting symptom. Hyperglycemia and diuresis produce symptoms of dehydration and may progress to hypovolemic shock. Metabolic acidosis causes rapid and deep respirations (Kussmaul sign). Impaired mentation is also seen in 10% of DKA patients and may quickly proceed to coma, especially if the serum osmolality exceeds 340 mOsm/kg.
Treatment
Fluid Replacement
Some controversy exists on how best to administer fluids in patients with DKA. In patients without renal impairment or shock, low-rate saline infusions of 500 ml/hr over 4 hours and then 250 ml/hr were more effective than more rapid rates of infusion. However, in the setting of hypotension, saline should be given more rapidly, and treatment should be guided by using central venous pressure measurements. Generally speaking, rates of 500 to 1,000 ml/hr of normal saline for the first 1 to 2 hours have been used to expand circulatory volume, followed by 250 to 500 ml/hr of normal saline or half-normal saline. It is generally accepted that half the estimated fluid deficit should be corrected in the first 12 hours. Dextrose and water should be added after glucose levels decrease to 250 mg/dl (14 mmol/l).
Insulin
With equivalent doses, intravenous insulin was shown to be as effective as subcutaneous or intramuscular insulin in reducing length of hospital stay, but intravenous insulin led to a faster initial decline in glucose and had more predictable effects [50]. Intravenous insulin should be administered early to reduce hospital stay and to ensure faster recovery. Low-dose insulin (e.g., 0.1 U/kg) is as effective as high-dose bolus therapy of 50 to 150 U and is associated with reduced instances of hypoglycemia and hypokalemia [51]. Therefore, the recommendation is for low-dose insulin bolus intravenously followed by maintenance therapy.
Maintenance Insulin
Intravenous insulin at a rate of 0.1 to 0.2 U/kg/hr is maintained until acidosis is mostly corrected (pH > 7.3, bicarbonate >18 mEq/l, or anion gap <14 mEq/l). Frequently, serum glucose decreases to less than 250 mg/dl (14 mmol/l) before acidosis is corrected. Insulin infusion should continue; however, a dextrose-and-water infusion should be started to prevent hypoglycemia. This practice of maintaining intravenous insulin is mainly to allow ketone clearance and correction of acidosis [52].
Bicarbonate
Controversy remains about the use of intravenous bicarbonate in DKA. No clear benefits were shown with intravenous bicarbonate administration with patients with pH that exceeds 6.9 [53, 54], and it can be associated with adverse events of hypokalemia and cerebral edema in children [55]. Therefore, bicarbonate use is not recommended. No studies have addressed the use of bicarbonate in patients with severe acidosis (pH < 6.9) or hypotension. In these patients, the practice has been to administer bicarbonate in doses of 44 to 133 mEq. No evidence supports or opposes such a practice.
Potassium
Potassium levels decrease quickly with management of acidosis because of intracellular compartment shift. Therefore, if serum potassium levels are low or normal, intravenous potassium should be administered immediately. If potassium levels are elevated (>5.5 mEq/l), potassium administration could be delayed until levels decrease to less than 5.5 mEq/l and the patient can urinate. Potassium is administered in 20 to 40 mEq/l of fluid provided as potassium chloride.
Phosphate
Serum phosphate levels also decreased to normal levels with proper management of DKA. Potential risk is associated with moderate to severe hypophosphatemia, including rhabdomyolysis, hemolysis, and impaired cardiac function. However, these are rarely found clinically. In small studies, adding phosphate to intravenous fluid did not affect the rate of recovery and was associated with minimal and nonsignificant lower serum calcium levels [56]. Therefore, phosphate treatment is not recommended unless evidence is found of significant hypophosphatemia.
Finally, in treating DKA, special attention should be directed to identifying as well as treating the triggering factors (e.g., urine culture, chest radiograph, and electrocardiogram).
Monitoring Therapy
Most patients with DKA experience hyperchloremic acidosis with therapy, especially in the first 8 hours. Therefore, observing serum bicarbonate levels, anion gap, and pH is better than measuring any single parameter. Correction of two of the three parameters (bicarbonate >18, pH > 7.3, and anion gap <14 mEq/l) is considered an adequate target of therapy. Assessing levels of urinary ketones as measured by standard assays measures only acetone and acetoacetate but not the primary ketone formed in DKA, which is β-hydroxybutyrate, and therefore, management of urine ketones should not be a target of therapy. As DKA is treated, β-hydroxybutyrate is converted to acetoacetate as well as acetone, both of which are readily measured by urinalysis, and therefore, excessive urine ketones can be seen, whereas serum levels of β-hydroxybutyrate have attained normal levels.
HYPEROSMOLAR COMA
Definition
Hyperosmolar coma is defined as extreme hyperglycemia that is associated with hyperosmolality, dehydration, and altered mental status without overt ketosis.
Etiology
A serious infection or an acute illness usually precipitates hyperosmolar coma in patients with diabetes. For many patients, hyperosmolar coma is the first manifestation of type 2 diabetes and may be related to severe dehydration and lack of access to drinking water [60]. Noncompliance with insulin treatment and surgical trauma are other important precipitating factors [61].
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
